Abstract
Diet and activity pattern changes consequent to urbanization are contributing to the global epidemic of cardiovasculardisease; less research has focused on activity within rural populations. We studied 527 women and 360 men (25–42 y), all rural-born and currently residing in rural or urban areas of Guatemala. We further classified rural male occupations as agricultural or nonagricultural. Overweight status (BMI ≥25 kg/m2) differed by residence/occupation among men (agricultural-rural, 27%; nonagricultural-rural, 44%; and urban, 55%; P < 0.01) and women (rural, 58%; and urban, 68%; P = 0.04). A moderate-to-vigorous lifestyle was reported by 76, 37, and 20% of men (agricultural-rural, nonagricultural-rural, and urban, respectively; P < 0.01); most women were sedentary, with no difference by residence. Prevalence of the metabolic syndrome was 17, 24, and 28% in agricultural-rural, nonagricultural-rural, and urban men, respectively (P = 0.2), and 44 and 45% in rural and urban women (P = 0.4). Dietary variables were largely unassociated with adiposity or cardio-metabolic risk factors; physical activity was inversely associated with the percentage of body fat in men. Percent body fat was inversely associated with HDL-cholesterol, and positively associated with triglycerides, blood pressure, and the metabolic syndrome in both men and women, and with LDL-cholesterol and fasting glucose in women. Differences in physical activity level, mainly attributable to occupation, appear more important than residence, per se, in influencing the risk for cardiovascular disease among men; differences among these sedentary women are likely related to other factors associated with an urban environment.
Introduction
The prevalence of cardiovascular disease (CVD)6 is increasing in low- and middle-income countries (1). In Guatemala, for example, mortality from CVD nearly doubled between 1986 and 1999, from 7 to 13% of all deaths, and currently CVD is the second leading cause of death (2). Overall, deaths due to CVD in nonwestern countries tend to occur a decade or 2 earlier than in western countries (3), contributing to estimates that disability-adjusted life years lost as a result of CVD are 2.8 times higher in developing countries than in developed countries (1).
In developing countries, the CVD epidemic is a consequence of long-term demographic and epidemiologic transitions (4), as well as the rapid changes in socio-cultural factors, collectively termed the “nutrition transition” (5). This transition is characterized by increased energy intake, mainly through fat and sugar consumption (6), decreased energy expenditure (7), and an increased prevalence of overweight and obesity (8,9). Rural-to-urban migration has been associated with the adoption of the lifestyle risk factors that comprise the nutrition transition, and the consequent development of metabolic CVD risk factors (10-12).
We have been studying a cohort of men and women born in 4 villages in Guatemala for >3 decades. We previously reported that, as young adults (19–29 y), urban migration was associated with increased adiposity and worsening lipid profiles in men, and with undesirable changes in diet and increased sedentarism in both sexes (13). This article updates and extends those findings. In the 1990s, road construction and improvements in transportation made larger municipalities more accessible (14), potentially blurring some of the distinctions between rural and urban environments. Furthermore, rural employment patterns have evolved; nearly all men in the 1960s and 1970s reported agricultural labor, compared with less than half in 2002 (14). We therefore considered both residence and primary occupation.
Methods
We examined 527 women and 360 men, who were born in 1 of the 4 rural study villages that participated in the Institute of Nutrition of Central America and Panama (INCAP) Longitudinal Study (1969–1977) (15) and who were resurveyed in the 2002–04 Human Capital Study. The target population for the resurvey included all individuals who had participated as children in the original study (n = 2392); among these, 274 had died (primarily due to infectious diseases at a young age), 162 were living out of the country, and 101 were untraceable. We attempted to contact the remaining 1855 persons. Completion rates were higher for instruments requiring only an interview (80% completed the food frequency and physical activity questionnaires) than for physical measures (73% completed the clinical exam, and 65% provided a fasting capillary blood sample) (4). Interviews of village residents were conducted by trained fieldworkers in the respondents' homes, except for the physical exam, which was conducted by a physician at local INCAP headquarters. For participants living elsewhere in Guatemala, data collection occurred either at INCAP headquarters in Guatemala City or during a home visit by a team composed of 2 interviewers and a physician. All data collection was approved by the human subjects review boards at both INCAP and Emory University, and informed consent was obtained from all participants.
Anthropometry
Height, weight, and abdominal circumference measures were taken in duplicate by trained interviewers. If there was a difference between the 2 measurements of >0.5 kg for body weight, 1.0 cm for height, or 1.5 cm for abdominal circumference, a third measure was taken and the closest 2 measures were used. We categorized participants as overweight (BMI ≥25 kg/m2) or obese (BMI ≥30 kg/m2) by BMI, and as having central obesity if the abdominal circumference was >102 cm (men) or <88 cm (women) (16). We calculated percent body fat using predictive equations derived from a similar population (17). Excess adiposity and obesity by percent body fat were defined as >24% and 30% (men) and >29% and >35% (women), respectively (18).
Physical activity
An interviewer asked participants about all activities performed on a typical work day, over the preceding year, including time spent sleeping and in various modes of transportation, primary and secondary occupations, chores, and leisure time activities. When reported activities did not sum to 24 h (reported mean ± SD was 1415.0 ± 86 min), we accounted for excesses by prorating activities and categorized any deficit as time spent in sedentary activities. We multiplied time in each activity by its respective metabolic equivalent (MET), which is a multiple of the basal metabolic rate (19), and determined physical activity levels by averaging MET-h for all activities over 24-h, or all activities other than sleeping over the waking time. In accordance with WHO criteria, we classified a moderate to vigorous lifestyle as a daily 24-h physical activity level ≥1.70 MET (20).
Diet
An interviewer-administered FFQ was used to ascertain typical consumption of 52 foods and beverages (including alcohol) over the previous 3 mo, as well as an open-ended section to allow for the inclusion of seasonal or less widely consumed foods. Nutrient intakes were estimated using the INCAP Nutrient Database (21) and supplemented with data from the U.S. Nutrient Database (22). When compared with 3 nonconsecutive 24-h recalls, the FFQ gives good measures of energy and macronutrient intake, as well as reasonably reliable measures of micronutrient intake (23).
Plasma lipids and glucose
Finger prick blood samples were collected after an overnight fast and analyzed with an enzymatic peroxidase dry chemistry method (Cholestech LDX System) to determine plasma total cholesterol (TC), HDL cholesterol (HDL-C), triglyceride, and glucose concentrations. In a previous study, Cholestech plasma lipid measures were compared with venous plasma collected at the time of the finger prick and analyzed at Emory University's Lipid Research Laboratory (24); triglycerides and HDL-C were similar, whereas TC tended to be underestimated by Cholestech. We used the Friedwald equation to determine LDL cholesterol (LDL-C) (25), except when triglycerides exceeded 400 mg/dL [(4.52 mmol/L) n = 25 persons in our sample]. We classified participants as having impaired fasting glucose when it was 100–125 mg/dL (5.6–7.0 mmol/L) (26), and as having diabetes when fasting glucose was ≥126 mg/dL (7.0 mmol/L) (27). Dyslipidemias were defined as TC ≥200 mg/dL (5.2 mmol/L), HDL-C <50 mg/dL (1.3 mmol/L) for women and <40 mg/dL (1.0 mmol/L) for men, LDL-C ≥130 mg/dL (3.4 mmol/L), triglycerides ≥150 mg/dL (1.7 mmol/L) (28), and a TC:HDL-C ratio ≥5 (29).
Blood pressure
Three blood pressure measurements were taken at least 3 min apart, after sitting quietly for at least 5 min. Measurements were taken with a digital sphygmomanometer (model UA-767, A&D Medical) on the left arm resting on a table at heart level. Three cuff sizes were available and were selected for use on the basis of arm circumference. If blood pressure measurements differed by >10 mm Hg, a fourth was taken; otherwise the second and third measures were recorded. We used the criteria for the metabolic syndrome to categorize elevated blood pressure (systolic ≥130 and/or diastolic blood pressure ≥85 mm Hg) (30).
Metabolic syndrome
We used the AHA/National Heart, Lung, and Blood Institute criteria for the metabolic syndrome (30). This included having 3 or more of the following 5 risk factors: central obesity, low HDL-C, elevated triglycerides, impaired fasting glucose, or elevated blood pressure.
Smoking
Participants identified themselves as current smokers or nonsmokers. Smokers were asked to report daily quantity of cigarettes smoked.
Residence/occupation
We classified participants living in 1 of the original 4 study villages as rural (n = 542) and those living in Guatemala City as urban (n = 174). For participants elsewhere in Guatemala (n = 171), we used household characteristics to categorize their residence as rural or urban. Rural men were further classified as agricultural-rural, if they described their primary occupation as farming or agricultural work, and nonagricultural-rural if they reported some other primary occupation. Eighty-three percent of rural women reported “head of household” as their primary occupation, and an additional 6% reported home-based commercial activities, including laundry or selling fruit or palm weavings; thus there was insufficient variation to stratify rural women by occupation.
Statistical methods
We restricted analysis to the 1063 persons who provided complete dietary, physical activity, anthropometric, and clinical data. We further excluded persons who reported fasting <9 h prior to blood sample collection (n = 127) or who reported an energy intake of >6000 kcal (25.1 MJ; n = 6) and women who were pregnant or lactating (n = 43). Excluded participants were more likely to be male (58 vs. 41%, P < 0.01) and have slightly higher triglycerides [209.5 ± 149 vs. 170.2 ± 88 mg/dL (2.37 ± 1.7 vs. 1.92 ± 1.0 mmol/L), P < 0.01], TC [167.9 ± 39 vs. 161.6 ± 32 mg/dL (4.36 ± 1.0 vs. 4.20 ± 0.8 mmol/L), P = 0.04], and HDL-C concentrations [39.1 ± 12 vs. 37.0 ± 10 mg/dL (1.0 ± 0.3 vs. 0.97 ± 0.3 mmol/L), P = 0.02]. There were no differences in TC:HDL-C ratio. We conducted analyses stratified by sex and residence/ occupation. We determined group differences by logistic regression and ANOVA with Tukey multiple comparison tests, all adjusted for age. To assess the association of lifestyle variables and cardio-metabolic risk factors, we used linear regression for the dependent variables, percent body fat, HDL-C, LDL-C, triglycerides, systolic and diastolic blood pressure, and glucose, and logistic regression for metabolic syndrome. All models were controlled for age and residence/occupation; dietary predictor variables were energy-adjusted using the residual method (31).
Results
Study participants were predominantly rural, and age 25–42 y at interview. Abdominal obesity was uncommon among men, despite a 43% prevalence of overweight. Among women, the prevalence of both overweight and abdominal obesity was close to 60%. BMI, percent body fat, abdominal circumference, and the prevalence of excess fat, overweight status, and obesity (by percent body fat) were all lowest in agricultural-rural men and highest in urban men (Table 1). Nonagricultural-rural men were generally similar to urban men with respect to anthropometric measures. Among women, there was a high prevalence of overweight and obesity (by BMI and percentage body fat) in both residence groups. BMI, percent body fat, and abdominal circumference were similar between groups; however differences were apparent in a higher prevalence of overweight and central obesity in urban women.
Table 1.
Selected characteristics of men and women surveyed in Guatemala, by sex and residence1
| Men |
Women |
||||
|---|---|---|---|---|---|
| Agricultural rural | Nonagricultural rural | Urban | Rural | Urban | |
| n | 88 | 153 | 119 | 372 | 155 |
| Age, y | 31.7 ± 4.4a | 31.4 ± 4.2a | 33.6 ± 4.3b | 32.6 ± 4.4 | 32.9 ± 4.0 |
| Height, cm | 162.1 ± 6.1 | 162.9 ± 6.1 | 163.8 ± 5.4 | 150.3 ± 5.4a | 151.8 ± 5.8b |
| BMI, kg/m2 | 23.6 ± 2.7a | 24.9 ± 3.7b | 25.5 ± 3.7b | 26.7 ± 5 | 27.1 ± 4 |
| Overweight (≥25 kg/m2), % | 27.3a | 43.8b | 54.6b | 58.1a | 67.7b |
| Obese by BMI (≥30 kg/m2), % | 3.4 | 9.6 | 11.8 | 25.3 | 20.0 |
| Body fat, % | 18.9 ± 5.1a | 20.9 ± 7.2ab | 22.2 ± 7.2b | 34.6 ± 7 | 35.5 ± 7 |
| Excess adiposity (>24% M, >30% W),2 % | 18.2a | 32.7b | 45.4b | 71.5 | 79.4 |
| Obese by body fat (>29% M, >35% W),2 % | 2.3a | 12.4b | 19.3b | 43.8 | 52.9 |
| Abdominal circumference, cm | 84.3 ± 6.7a | 87.4 ± 9.8b | 89.3 ± 9.9b | 91.7 ± 12.1 | 93.1 ± 10.8 |
| Abdominal obesity (>102 M, >88 W),2 % | 0a | 8.5b | 8.4b | 58.9a | 68.4b |
| Energy intake,3 kcal/d | 3278 ± 926b | 3146 ± 788b | 2850 ± 834a | 2329 ± 729b | 2154 ± 633a |
| Energy intake, kcal · kg−1 · d−1 | 53.2 ± 15.0c | 48.6 ± 14.2b | 42.4 ± 13.4a | 40.0 ± 14.4b | 35.5 ± 12.2a |
| Energy from fat, % | 17.4 ± 4.5 | 18.5 ± 3.9 | 18.7 ± 5.1 | 18.2 ± 4.9a | 19.5 ± 4.9b |
| Energy from carbohydrates, % | 70.7 ± 5.2 | 69.5 ± 4.6 | 69.2 ± 5.6 | 69.6 ± 5.2b | 68.1 ± 6.2a |
| Fiber, g · 1000 kcal−1 · d−1 | 13.9 ± 2.4b | 12.7 ± 2.0a | 12.6 ± 2.2a | 12.8 ± 2.4b | 12.2 ± 2.6a |
| Cholesterol, mg · 1000 kcal−1 · d−1 | 112.7 ± 70.2a | 127.3 ± 60.3a | 153.5 ± 80.3b | 118.0 ± 68.7a | 131.8 ± 70.8b |
| Physical activity level | |||||
| 24-h, MET/d | 1.81 ± 0.2c | 1.64 ± 0.3b | 1.55 ± 0.3a | 1.43 ± 0.1a | 1.46 ± 0.1b |
| Waking, MET/d | 2.74 ± 0.3c | 2.48 ± 0.5b | 2.31 ± 0.4a | 2.22 ± 0.2 | 2.21 ± 0.2 |
| ≥1.7, % | 76.4c | 37.2b | 20.2a | 3.5 | 4.5 |
| Current smoker, % | 46.1ab | 44.9b | 30.3a | 1.8 | 0.6 |
Values are means ± SEM or %. Within a sex, values in a row with superscripts without a common letter differ, P < 0.05.
M, men; W, women.
1 kcal = 4.184 kJ.
The diet was characterized by low fat (17–20% of energy) and high carbohydrate (68–71% of energy) intakes (32); dietary fiber intake was high (32). There were minor differences among groups in macronutrient composition of the diet. Rural men (both agricultural and nonagricultural) had a higher total energy intake and lower cholesterol intake than urban men, whereas fiber intake was higher in agricultural-rural men than in both nonagricultural-rural and urban men. Energy intake independent of body weight was highest in agricultural-rural men and lowest in urban men, with differences across all 3 groups. Similarly, compared with urban women, rural women had a higher energy and fiber intake and lower cholesterol and fat intake.
Agricultural-rural men were significantly more active (76.4% had a physical activity level >1.70) than nonagricultural-rural or urban men (37.2 and 20.2%, respectively; P < 0.01). Women were predominantly sedentary, with only ∼4% categorized as having a physical activity level > 1.70. Physical activity level was slightly lower for women living in rural than in urban areas (P = 0.02). Smoking was more likely to be reported by rural than urban men and was rare among women. Among those who did report smoking, mean smoking frequency was low, <2 cigarettes/d.
Overall, the metabolic profile of this population was characterized by low HDL-C and elevated triglycerides (Table 2). Low HDL-C was prevalent in 75% of men and 87% of women (P < 0.01), with no significant differences by residence. Triglyceride levels were >150 mg/dL (1.7 mmol/L) in close to 50% of both men and women, with no differences by residence among men. In women, there were few differences in lipid profiles due to residence. The TC:HDL-C ratio exceeded 5.0 in one-half of the urban men, compared with one-third of agricultural-rural men (P < 0.01). Among men, elevated blood pressure tended to be higher in urban residents (P = 0.06). Among women, systolic and diastolic blood pressure, and the prevalence of elevated blood pressure, were higher in those living in urban areas. There were no differences by residence in glucose measures among men or women. Prevalence of the metabolic syndrome was lower in men than women (23 vs. 44%, respectively, P < 0.01), and there was no difference by residence.
Table 2.
Selected cardio-metabolic risk factors of men and women surveyed in Guatemala in 2002–2004, by sex and residence1
| Men |
|||||
|---|---|---|---|---|---|
| Agricultural rural |
Nonagricultural rural |
Urban | Women |
||
| Rural | Urban | ||||
| n | 88 | 153 | 119 | 372 | 155 |
| Total cholesterol,2 mg/dL | 153.2 ± 31.3a | 155.3 ± 32.2a | 168.4 ± 35.1b | 162.7 ± 30.4 | 164.8 ± 32.7 |
| ≥200, % | 11.4 | 10.5 | 17.7 | 12.1 | 13.6 |
| HDL-C, mg/dL | 36.4 ± 10.4b | 32.7 ± 8.1a | 33.4 ± 8.8ab | 38.7 ± 10.8 | 40.1 ± 10.0 |
| <40 M, <50 W,3 % | 65.9 | 79.1 | 76.5 | 87.4 | 86.5 |
| Total:HDL-C ratio | 4.5 ± 1.6a | 5.0 ± 1.5ab | 5.4 ± 1.9b | 4.5 ± 1.3 | 4.3 ± 1.3 |
| ≥5, % | 30.7a | 46.4ab | 52.9b | 30.7 | 23.2 |
| LDL-C, mg/dL | 84.0 ± 26.5a | 88.1 ± 26.5a | 99.6 ± 31.7b | 90.1 ± 25.2 | 93.7 ± 27.4 |
| ≥130, % | 5.7 | 8.2 | 13.4 | 5.5a | 10.1b |
| Triglycerides, mg/dL | 162.9 ± 72.7 | 176.2 ± 96.7 | 182.8 ± 101.8 | 169.1 ± 81.0 | 161.6 ± 89.2 |
| ≥150, % | 46.6 | 54.9 | 54.6 | 53.2b | 43.9a |
| Systolic blood pressure, mm Hg | 114.9 ± 9.8 | 116.7 ± 11.5 | 118.1 ± 11.6 | 107.2 ± 11.6a | 111.9 ± 15.6b |
| Diastolic blood pressure, mm Hg | 71.3 ± 7.0 | 72.2 ± 9.5 | 72.8 ± 10.0 | 69.3 ± 8.6a | 72.4 ± 11.0b |
| SBP ≥130 or DBP ≥85, % | 11.4 | 15.7 | 23.5 | 6.5a | 14.2b |
| Plasma glucose, mg/dL | 91.6 ± 8.6 | 92.7 ± 9.7 | 92.0 ± 10.6 | 94.5 ± 24.8 | 97.0 ± 40.0 |
| Impaired fasting glucose 100–125, % | 20.5 | 18.3 | 17.7 | 19.4 | 15.5 |
| Diabetes mellitus ≥126, % | 0 | 1.3 | 0.8 | 2.2 | 4.5 |
| Metabolic syndrome, % | 17.1 | 23.5 | 27.7 | 44.1 | 44.5 |
Values are means ± SEM or %. Within a sex, values in a row with superscripts without a common letter differ, P < 0.05.
Conversions: mg/dL cholesterol to mmol/L, multiply by 0.026; mg/dL triglycerides to mmol/L, multiply by 0.0113; mg/dL glucose to mmol/L, multiply by 0.056.
M, men; W, women.
We found few associations among energy intake and energy-adjusted intakes of fat, carbohydrate, fiber, or cholesterol and percent body fat or cardio-metabolic risk factors in men or women (Table 3). There was a slight association between carbohydrate and percent body fat in men and women, although in opposite directions. In women, dietary cholesterol intake was associated inversely with percent body fat, and positively with fasting glucose. In men, physical activity level was inversely associated with percent body fat. Among both men and women, percent body fat was inversely associated with HDL and positively associated with triglycerides, systolic and diastolic blood pressure, and the metabolic syndrome. In women, percent body fat was also positively associated with LDL and fasting glucose.
Table 3.
Associations of diet, physical activity, and adiposity with cardio-metabolic risk factors in men and women surveyed in Guatemala in 2002–2004, by sex1
| Body fat |
HDL-C |
LDL-C |
Triglycerides |
Systolic blood pressure |
Diastolic blood pressure |
Glucose |
Metabolic syndrome |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β | P | β | P | β | P | β | P | β | P | β | P | β | P | β | P | |
| Men, n = 360 | % | mg/dL2 | mg/dL2 | mg/dL2 | mm Hg | mm Hg | mg/dL2 | |||||||||
| Dietary energy,3 kcal/d | 0.52 | 0.2 | 0.28 | 0.6 | −3.23 | 0.07 | 4.18 | 0.5 | 0.97 | 0.2 | 0.67 | 0.2 | 0.10 | 0.9 | −0.07 | 0.7 |
| Dietary fat,4 g/d | 0.03 | 0.1 | −0.01 | 0.6 | 0.11 | 0.2 | 0.21 | 0.5 | −0.01 | 0.7 | −0.002 | 0.9 | 0.03 | 0.3 | 0.003 | 0.7 |
| Dietary carbohydrate,4 g/d | −0.02 | 0.03 | 0.002 | 0.8 | −0.02 | 0.6 | −0.11 | 0.3 | −0.004 | 0.8 | −0.002 | 0.9 | −0.01 | 0.3 | −0.003 | 0.4 |
| Dietary fiber,4 g/d | −0.02 | 0.7 | 0.08 | 0.2 | 0.07 | 0.7 | −1.21 | 0.08 | 0.05 | 0.6 | 0.003 | 0.9 | −0.02 | 0.8 | −0.02 | 0.5 |
| Dietary cholesterol,4 mg/d | 0.002 | 0.2 | −0.004 | 0.1 | 0.01 | 0.2 | 0.02 | 0.4 | 0.009 | 0.8 | 0.002 | 0.4 | 0.001 | 0.6 | 0.004 | 0.6 |
| Physical activity level, MET | −5.94 | <0.001 | 1.54 | 0.4 | −7.31 | 0.2 | −16.30 | 0.4 | −2.16 | 0.3 | −2.68 | 0.1 | −0.88 | 0.7 | −0.9 | 0.1 |
| Adiposity, % body fat | n/a | n/a | −0.47 | <0.001 | −0.09 | 0.7 | 5.08 | <0.001 | 0.69 | <0.001 | 0.50 | <0.001 | 0.09 | 0.3 | 0.19 | <0.001 |
| Women, n = 527 | ||||||||||||||||
| Dietary energy,3 kcal/d | 0.68 | 0.1 | −0.28 | 0.7 | 1.48 | 0.4 | 2.69 | 0.6 | 1.50 | 0.06 | 0.74 | 0.2 | −0.42 | 0.8 | 0.11 | 0.4 |
| Dietary fat,4 g/d | −0.04 | 0.07 | 0.02 | 0.7 | −0.02 | 0.8 | 0.07 | 0.8 | −0.07 | 0.1 | −0.02 | 0.5 | 0.008 | 0.2 | −0.002 | 0.8 |
| Dietary carbohydrate,4 g/d | 0.02 | 0.03 | −0.01 | 0.3 | 0.01 | 0.7 | 0.03 | 0.8 | 0.02 | 0.1 | 0.007 | 0.6 | −0.008 | 0.07 | 0.002 | 0.5 |
| Dietary fiber,4 g/d | −0.01 | 0.9 | 0.02 | 0.7 | −0.12 | 0.6 | −0.66 | 0.3 | 0.03 | 0.8 | −0.04 | 0.5 | 0.05 | 0.5 | −0.01 | 0.4 |
| Dietary cholesterol,4 mg/d | −0.004 | 0.03 | 0.003 | 0.3 | 0.003 | 0.7 | −0.02 | 0.4 | −0.006 | 0.1 | −0.003 | 0.2 | 0.007 | 0.02 | −0.004 | 0.5 |
| Physical activity level, MET | −2.21 | 0.5 | 0.23 | 0.9 | −13.29 | 0.3 | −49.89 | 0.2 | 6.78 | 0.2 | −0.35 | 0.9 | 13.46 | 0.3 | −0.03 | 0.9 |
| Adiposity, % body fat | n/a | n/a | −0.45 | <0.001 | 0.49 | 0.002 | 3.28 | <0.001 | 0.78 | <0.001 | 0.42 | <0.001 | 0.56 | 0.002 | 0.20 | <0.001 |
Adjusted for age and residence/occupation.
Conversions: mg/dL cholesterol to mmol/L, multiply by 0.026; mg/dL triglycerides to mmol/L, multiply by 0.0113; mg/dL glucose to mmol/L, multiply by 0.056.
1 kcal = 4.184 kJ.
Intakes adjusted for energy intake by the residual method.
Discussion
We compared the lifestyle and metabolic CVD risk factors of men and women who shared a similar early-life rural environment and currently reside in rural or urban areas of Guatemala. Our a priori hypotheses were that adoption of less-desirable dietary habits by urban residents, and decreased levels of physical activity in nonagricultural-rural men and urban residents, would contribute to increased adiposity and cardio-metabolic risk. We observed few associations of dietary variables with percent body fat or cardio-metabolic risk factors, and we found physical activity to be inversely associated with percent body fat in men, but not women. Percent body fat was associated with most metabolic and cardiovascular risk factors in men and women, but physical activity level was not, suggesting the effects of activity on CVD risk in this population are primarily mediated by body composition.
We considered BMI, percentage body fat, and abdominal circumference as independent measures of body composition. Results were generally consistent with all 3 measures, and thus, we report only data for percent body fat. The observed associations between percent body fat and physical activity emphasize the importance of physical activity, rather than dietary intake, on body composition. The importance of physical activity is also evident in rural men. Men with traditional agricultural occupations had a more desirable cardio-metabolic profile, suggesting that the opportunities for physical activity created by different environments are likely to be a significant factor contributing to associations between urbanization and CVD risk. A study in rural and urban Cameroon found higher energy and fat intake, but lower BMI, blood pressure, and total cholesterol in rural residents (33). Similarly, a study in 3 Pacific island countries found higher energy and carbohydrate intake, but lower prevalence of diabetes and hypertension in rural residents (34). Both studies attributed the lower levels of obesity and CVD risk factors to the significantly higher physical activity levels in the rural population.
We observed few associations between dietary variables and cardio-metabolic risk factors. Data were subject to respondent bias due to self-reporting, and to imprecision in our instrument. Our FFQ was developed among a rural population (23), so there may be aspects of the urban diet that were not adequately captured. We did not collect data on added salt and did not have sufficient data for analyzing specific fatty acid components, which may have differential effects on blood pressure and the lipid profile. Additionally, there may be insufficient variance in dietary intakes in this population, independent of variation in energy intake consequent to variation in physical activity, to detect modest associations. In other settings, a high carbohydrate diet has been shown to be an underlying cause of both elevated triglycerides and low HDL-C (28,35), and this is likely contributing to the high prevalence of both, as well as the metabolic syndrome, such as the 17% of agricultural-rural men.
We found rural-urban differences in physical activity in men but not women. Our physical activity questionnaire was developed for this population to collect detailed information on various domains of activity (occupation, transportation, chores, etc.), which contribute significantly to overall physical activity levels in developing countries. Physical activity research has largely focused on leisure time, rather than multiple domains of daily activity; such strategies have been shown to lead to biased estimates in developing countries (36,37). Despite the inclusion of various domains of activity, most women reported “head of household” as their primary occupation, which likely masks a range of intensity levels that we are not able to discriminate, depending upon household amenities and number of children at home.
Our previous analysis of data collected in this cohort in 1997 (13) found no differences in blood pressure between rural and urban women, although we found higher systolic and diastolic blood pressures, and prevalence of elevated blood pressure, in urban women. Several factors may be contributing to these differences, including age, significantly higher prevalence of obesity, and potential for longer time spent in an urban environment in those included in this analysis. There may also be differences in the subsets of the cohort that were examined at each wave. This study includes many individuals not contacted for the 1997 survey; those contacted for the 2002–04 survey tend to be older and are more likely to live in locations in Guatemala other than the original villages or Guatemala City.
Our study population represents a cohort that we have been following prospectively for > 3 decades. Few such cohorts exist, and we have a unique opportunity to study the effects of migration in a population that shares an early-life rural environment. Nevertheless, participants included in this analysis were not a random sample of the original study cohort. An analysis of attrition found that, although there were some differences in the groups not interviewed, there were few differences when comparing all noninterviewed participants with interviewed participants, suggesting no systematic bias (38). Additionally, we found few differences between participants with data that were included or excluded from this analysis.
Others have shown recent urbanization, cumulative exposure to an urban environment, or number of relatives in urban centers to be independently associated with overweight and obesity (12,39). Our rural-urban classifications were based solely on current residence; we did not measure duration of time living in an urban environment or any other indicator of exposure to an urban environment. Cyclical migration patterns may also attenuate rural-urban differences. In the past several decades, highway access roads have been built, dirt roads have been paved, and bus services now run more frequently from all of the villages to larger municipalities, creating much greater interaction between rural and urban environments (14).
A higher prevalence of detrimental health behaviors and outcomes associated with urban compared with rural environments has been well documented; however there are inconsistencies by sex, indicators of urbanization, and conclusions regarding the most significant contributors to CVD risk (10,12,13,40-42). Our analysis is generally consistent with findings that urban residence is associated with increased CVD risk; however our further stratification of rural men by occupation, and that we found fewer differences among women, suggest that it is not simply where participants are residing, but the activities and behaviors they are performing, that are important. A multicountry analysis found that, at increasing levels of economic productivity, likely associated with more sedentary and professional occupations in rural environments, rural-urban discrepancies in overweight and obesity decreased (43). Thus, sedentarization of rural occupations needs to be considered in addition to previously documented changes consequent of rural to urban migration.
Footnotes
Supported by NIH TW005598 and HD046125. Cria O. Gregory and Jun Dai are both supported by predoctoral fellowship awards from the AHA.
Author disclosures: C. O. Gregory, no conflicts of interest; J. Dai, no conflicts of interest; M. Ramirez-Zea, no conflicts of interest; and A. Stein, no conflicts of interest.
Abbreviations used: CVD, cardiovascular disease; HDL-C, HDL cholesterol; INCAP, Institute of Nutrition of Central America and Panama; LDL-C, LDL cholesterol; MET, metabolic equivalent; TC, total cholesterol.
Literature Cited
- 1.WHO . World health report 1999: Making a difference. World Health Organization; Geneva: 1999. [Google Scholar]
- 2.PAHO . Health situation in the Americas, basic indicators. Pan American Health Organization; Washington, D.C.: 2003. [Google Scholar]
- 3.Reddy KS. Cardiovascular disease in non-Western countries. N Engl J Med. 2004;350:2438–40. doi: 10.1056/NEJMp048024. [DOI] [PubMed] [Google Scholar]
- 4.Omran AR. The epidemiologic transition. A theory of the epidemiology of population change. Milbank Mem Fund Q. 1971;49:509–38. [PubMed] [Google Scholar]
- 5.Reddy KS, Yusuf S. Emerging epidemic of cardiovascular disease in developing countries. Circulation. 1998;97:596–601. doi: 10.1161/01.cir.97.6.596. [DOI] [PubMed] [Google Scholar]
- 6.Drewnowski A, Popkin BM. The nutrition transition: new trends in the global diet. Nutr Rev. 1997;55:31–43. doi: 10.1111/j.1753-4887.1997.tb01593.x. [DOI] [PubMed] [Google Scholar]
- 7.Popkin BM. Urbanization, lifestyle changes, and the nutrition transition. World Dev. 1999;27:1905–16. [Google Scholar]
- 8.Uauy R, Albala C, Kain J. Obesity trends in Latin America: transiting from under- to overweight. J Nutr. 2001;131:893S–9S. doi: 10.1093/jn/131.3.893S. [DOI] [PubMed] [Google Scholar]
- 9.Martorell R, Khan LK, Hughes ML, Grummer-Strawn LM. Obesity in women from developing countries. Eur J Clin Nutr. 2000;54:247–52. doi: 10.1038/sj.ejcn.1600931. [DOI] [PubMed] [Google Scholar]
- 10.He J, Klag MJ, Wu Z, Qian MC, Chen JY, Mo PS, He QO, Whelton PK. Effect of migration and related environmental changes on serum lipid levels in southwestern Chinese men. Am J Epidemiol. 1996;144:839–48. doi: 10.1093/oxfordjournals.aje.a009018. [DOI] [PubMed] [Google Scholar]
- 11.Carrasco EP, Perez FB, Angel BB, Albala CB, Santos JL, Larenas GY, Montalvo DV. Prevalence of type 2 diabetes and obesity in two Chilean aboriginal populations living in urban zones. Rev Med Chil. 2004;132:1189–97. doi: 10.4067/s0034-98872004001000005. [DOI] [PubMed] [Google Scholar]
- 12.Sobngwi E, Mbanya JC, Unwin NC, Porcher R, Kengne AP, Fezeu L, Minkoulou EM, Tournoux C, Gautier JF, et al. Exposure over the life course to an urban environment and its relation with obesity, diabetes, and hypertension in rural and urban Cameroon. Int J Epidemiol. 2004;33:769–76. doi: 10.1093/ije/dyh044. [DOI] [PubMed] [Google Scholar]
- 13.Torun B, Stein AD, Schroeder D, Grajeda R, Conlisk A, Rodriguez M, Mendez H, Martorell R. Rural-to-urban migration and cardiovascular disease risk factors in young Guatemalan adults. Int J Epidemiol. 2002;31:218–26. doi: 10.1093/ije/31.1.218. [DOI] [PubMed] [Google Scholar]
- 14.Maluccio JA, Melgar P, Mendez H, Murphy A, Yount KM. Social and economic development and change in four Guatemalan villages: demographics, schooling, occupation, and assets. Food Nutr Bull. 2005;26(2 Suppl 1):S25–45. doi: 10.1177/15648265050262S104. [DOI] [PubMed] [Google Scholar]
- 15.Martorell R, Habicht JP, Rivera JA. History and design of the INCAP longitudinal study (1969–77) and its follow-up (1988–89) J Nutr. 1995;125(4 Suppl):1027S–41S. doi: 10.1093/jn/125.suppl_4.1027S. [DOI] [PubMed] [Google Scholar]
- 16.WHO . Physical Status: The Use and Interpretation of Anthropometry. World Health Organization; Geneva: 1995. [Google Scholar]
- 17.Ramirez-Zea M, Torun B, Martorell R, Stein AD. Anthropometric predictors of body fat as determined by hydrostatic weighing in Guatemalan adults. Am J Clin Nutr. 2006;83:795–802. doi: 10.1093/ajcn/83.4.795. [DOI] [PubMed] [Google Scholar]
- 18.ACSM . ACSM's Guidelines for Exercise Testing and Prescription. 6th Ed. Williams & Wilkins; Baltimore, MD: 1998. [Google Scholar]
- 19.Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath SJ, O'Brien WL, Bassett DR, Jr., Schmitz KH, et al. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc. 2000;32(9 Suppl):S498–504. doi: 10.1097/00005768-200009001-00009. [DOI] [PubMed] [Google Scholar]
- 20.FAO/WHO/UNU . Human Energy Requirements: Report of Joint FAO/WHO/UNU Expert Consultation. FAO/WHO/UNU; Rome: 2004. [Google Scholar]
- 21.Menchu M, Mendez H, Barrera M, Ortega L. Nutritional value of Central American foods. Institute of Nutrition of Central America and Panama; Guatemala City: 1996. [Google Scholar]
- 22.USDA . USDA Nutrient Database for Standard Reference, release 12. United States Department of Agriculture; Washington, D.C.: 1998. [Google Scholar]
- 23.Rodriguez MM, Mendez H, Torun B, Schroeder D, Stein AD. Validation of a semi-quantitative food-frequency questionnaire for use among adults in Guatemala. Public Health Nutr. 2002;5:691–9. doi: 10.1079/PHN2002333. [DOI] [PubMed] [Google Scholar]
- 24.Flores R, Grajeda R, Torun B, Mendez H, Martorell R, Schroeder D. Evaluation of a dry chemistry method for blood lipid in field studies. FASEB J. 1998;12:3061. [Google Scholar]
- 25.Friedwald W, Levy R, Fredrickson D. Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499–502. [PubMed] [Google Scholar]
- 26.Grundy SM, Hansen B, Smith SC, Jr., Cleeman JI, Kahn RA. Clinical management of metabolic syndrome: report of the American Heart Association/National Heart, Lung, and Blood Institute/American Diabetes Association conference on scientific issues related to management. Circulation. 2004;109:551–6. doi: 10.1161/01.CIR.0000112379.88385.67. [DOI] [PubMed] [Google Scholar]
- 27.Genuth S, Alberti KG, Bennett P, Buse J, Defronzo R, Kahn R, Kitzmiller J, Knowler WC, Lebovitz H, et al. Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care. 2003;26:3160–7. doi: 10.2337/diacare.26.11.3160. [DOI] [PubMed] [Google Scholar]
- 28.NIH . Third Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) National Institutes of Health, National Heart, Lung, and Blood Institute; Bethesda, MD: 2001. [Google Scholar]
- 29.Wang TD, Chen WJ, Chien KL, Seh-Yi Su SS, Hsu HC, Chen MF, Liau CS, Lee YT. Efficacy of cholesterol levels and ratios in predicting future coronary heart disease in a Chinese population. Am J Cardiol. 2001;88:737–43. doi: 10.1016/s0002-9149(01)01843-4. [DOI] [PubMed] [Google Scholar]
- 30.Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/ National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112:2735–52. doi: 10.1161/CIRCULATIONAHA.105.169404. [DOI] [PubMed] [Google Scholar]
- 31.Willett W. Nutritional Epidemiology. 2nd ed. Oxford University Press; New York: 1998. [Google Scholar]
- 32.IOM . Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acds, Cholesterol, Protein, and Amino Acids. Institute of Medicine; Washington, D.C.: 2002. [DOI] [PubMed] [Google Scholar]
- 33.Mennen LI, Mbanya JC, Cade J, Balkau B, Sharma S, Chungong S, Cruickshanck JK. The habitual diet in rural and urban Cameroon. Eur J Clin Nutr. 2000;54:150–4. doi: 10.1038/sj.ejcn.1600909. [DOI] [PubMed] [Google Scholar]
- 34.Taylor R, Badcock J, King H, Pargeter K, Zimmet P, Fred T, Lund M, Ringrose H, Bach F, et al. Dietary intake, exercise, obesity and noncommunicable disease in rural and urban populations of three Pacific Island countries. J Am Coll Nutr. 1992;11:283–93. doi: 10.1080/07315724.1992.10718229. [DOI] [PubMed] [Google Scholar]
- 35.Liu S, Manson JE, Stampfer MJ, Holmes MD, Hu FB, Hankinson SE, Willett WC. Dietary glycemic load assessed by food-frequency questionnaire in relation to plasma high-density-lipoprotein cholesterol and fasting plasma triacylglycerols in postmenopausal women. Am J Clin Nutr. 2001;73:560–6. doi: 10.1093/ajcn/73.3.560. [DOI] [PubMed] [Google Scholar]
- 36.Tudor-Locke C, Neff LJ, Ainsworth BE, Addy CL, Popkin BM. Omission of active commuting to school and the prevalence of children's health-related physical activity levels: the Russian Longitudinal Monitoring Study. Child Care Health Dev. 2002;28:507–12. doi: 10.1046/j.1365-2214.2002.00295.x. [DOI] [PubMed] [Google Scholar]
- 37.Hu G, Pekkarinen H, Hanninen O, Yu Z, Tian H, Guo Z, Nissinen A. Physical activity during leisure and commuting in Tianjin, China. Bull World Health Organ. 2002;80:933–8. [PMC free article] [PubMed] [Google Scholar]
- 38.Grajeda R, Behrman JR, Flores R, Maluccio JA, Martorell R, Stein AD. The human capital study 2002–04: tracking, data collection, coverage, and attrition. Food Nutr Bull. 2005;26(2 Suppl 1):S15–24. doi: 10.1177/15648265050262S103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Ulijaszek SJ, Koziel S, Hermanussen M. Village distance from urban centre as the prime modernization variable in differences in blood pressure and body mass index of adults of the Purari delta of the Gulf Province, Papua New Guinea. Ann Hum Biol. 2005;32:326–38. doi: 10.1080/03014460500087824. [DOI] [PubMed] [Google Scholar]
- 40.Unwin N, McLarty D, Machibya H, Aspray T, Tamin B, Carlin L, Patel S, Walker M, Alberti KG. Changes in blood pressure and lipids associated with rural to urban migration in Tanzania. J Hum Hypertens. 2006;20:704–6. doi: 10.1038/sj.jhh.1002056. [DOI] [PubMed] [Google Scholar]
- 41.Abdul-Rahim HF, Husseini A, Bjertness E, Giacaman R, Gordon NH, Jervell J. The metabolic syndrome in the West Bank population: an urban-rural comparison. Diabetes Care. 2001;24:275–9. doi: 10.2337/diacare.24.2.275. [DOI] [PubMed] [Google Scholar]
- 42.Hodge AM, Dowse GK, Toelupe P, Collins VR, Imo T, Zimmet PZ. Dramatic increase in the prevalence of obesity in western Samoa over the 13 year period 1978–1991. Int J Obes Relat Metab Disord. 1994;18:419–28. [PubMed] [Google Scholar]
- 43.Mendez MA, Monteiro CA, Popkin BM. Overweight exceeds under-weight among women in most developing countries. Am J Clin Nutr. 2005;81:714–21. doi: 10.1093/ajcn/81.3.714. [DOI] [PubMed] [Google Scholar]